Any reason I would choose Unity over Unreal?

True, navigation in Blender is actually quite fast once you "grok" it... still, I mistype ctrl and shift sometimes when trying to move the view, as I am not looking at the keyboard.

With the space navigator, once you are used to this new method of navigation, that will not happen. Once your muscle memory kicks in, you navigate without thinking about it.
I use it for sculpting in 3D Coat (where I don't use a mouse, but a cintiq tablet), and it really speeds up the process.
For Blender, I do not use it that much, but then Blender needs a mouse and KB... don't know, should probably move the space navigator to the other side of the mouse and see if Mouse+Space Navigator is just as awesome as Pen+Space Navigator.

Sure, the Blender system of navigation in Unity would be like a 1000x better than what we have now. Still, having the Space Navigator Plugin is better as long as you have the hardware.


If you can (in a store or so), give it a try once. Even if you don't intend to buy it. You really need some time to get used to it, but after a while you will see what I am talking about.
Really a shame there is no competition and thus no cheaper 25$ SpaceNavigators around (it shouldn't be THAT much more expensive to produce than mice).
I am certain everyone who is working in 3D would fall in love with it IF it wasn't priced outside of their price range... and given how this is an "addon utility" for most people, and one of the things you need to try first before seeing the need for it, it is just a tad too expensive to be the runaway success it could be. Not everyone is spending 100$ for something that MIGHT be useful for them.

 

Hmm....
Checked youtube video, it is basically a mini-flightstick with 6 axes of movement (3 rotation + 3 shift) instead of 3 (rotational) all mapped to a single control.
I see how it would work, not convinced it is something I'd be using though.

Keyboard and tablet works well (for sculpting, anyway - moving points is easier with mouse), but then again, I don't have a cintiq.

 

Well, I was getting it after being a little bit irritated with navigation in multiple 3D apps (before I really "groked" Blenders weird UI)... and I was sceptical at first too (yeah, I felt some buyers remorse about the 100$ wasted on it).

But it grew on me, which is pretty much what every Space Navigator user I heard was telling on the net. Its quite a different way to interact with your 3D Canvas.


Anyway, not trying to sell anything here...

 

You can have a signature with images in it though.. and you can earn yourself a extra ribbons. Unity really should tie the answer hub/issue/feedback stuff into forum profiles. Unity should do a lot of things though.

"Basically the only two nice features C#"

The biggest nice feature of Java/C# is its a language that reads clearly and is well designed. Without verbose symbol << s << h << i << t syntax everywhere and extra header files that just add to the headache of using it. Its a language that has been dragged through the decades, where every decision on new features and language design improvements or even compiler improvements like nested block comments have been ignored because 'compatibility' with the stone age of a computing. I don't hate the language though it just seems to be a mess, where Java/C# gets so much right for todays developer and hardware, while performance and memory handling may never be as good, for the most part its good enough.

Anyway TheAlmightyPixel pro/con post sums up a lot.. and if you're coming from a background of scripting languages and Java/C# then learning and using Unity is super easy, and its large community and learning resources just adds to the ease of use. Though not in good ways I mean you'll come across tons of threads dating back months/years regarding stupid effing issues and bugs and forgotten feature requests that no one at unity can be bothered to add/fix. I mean just recently mine was wheretf is the build report regarding whattf is being shoved into these .asset bundles i can't natively open with unity or any external program it ships with.. .the answer was the build report for file sizes is no longer shown in the editor console after building, you'll find it buried somewhere in the unity editor log that's buried in user folder somewhere. A wise unity developer would probably have sorted out some sort of gui to visualize all this S***, or just show the build report in the console (supposedly it used to work that way according documentation).

And here is list of things Unreal did since....
June 2015
https://www.unrealengine.com/blog/unreal-engine-48-released
August 2015
https://www.unrealengine.com/blog/unreal-engine-49-released
November 2015
https://www.unrealengine.com/blog/unreal-engine-4-10-released
March 2016
https://www.unrealengine.com/blog/unreal-engine-4-11-released
June 2016
https://www.unrealengine.com/blog/unreal-engine-4-12-released

Unreal have been killing it terms of engine/feature improvements and its already got more than adequate base tools built in than Unity.

For Unity releases in the same timeframe..

June 2015
https://unity3d.com/unity/whats-new/unity-5.1
September 2015
https://unity3d.com/unity/whats-new/unity-5.2
December 2015
https://unity3d.com/unity/whats-new/unity-5.3
http://blogs.unity3d.com/2015/12/08/unity-5-3-all-new-features-and-more-platforms/

5.5 has 2d preview the rest still waiting, 5.4 probably not long now I suppose, but 5.4 doesn't really bring in great features, or even really any useful editor improvements.. it does bring that giant S*** heap that is dx12 that only works for win10 support that no one but M$ care about, oh and gamers who all jumped on that wreckage in the hopes of better fps.. how much disappoint can they take with that OS. And obviously lots of work put towards that effort, but still no vulkan support let alone on desktop albeit unreal only support it for the mobile officially but its further ahead than here. Unity cinema effects stuff was released in beta, none of it was any better that same Unreal built in effects. Chman ssao pro is better with more settings to tame the effect for a game.. but he now works at unity on what I imagine are cinema effects.. so its like Unity have set limits on what they will release to not interfere with the asset store stuff.. and that's fine, they've become overly reliant on it to me, and I can't see it really helping in the longterm when the overall quality and accessible feature set overshadowed by competitors. Yeh not in all areas, but sometimes you only care about a few things done really welll, but I'm sure they don't care until users do start switching.

So its releases for the past year. ..Bugs, lackluster features and delays. The way Unity was designed from the early days, its c# component system, the community, knowledge, more script resources, tutorials, and assets ...are its biggest asset... but how its going to hold up against competitors when they slip on so many areas isn't that good. The service stuff and ad platform are mainly mobile orientated features imo. And the features I thought they'd have out by now are somewhere in a roadmap still, things like nested prefabs should have been out way sooner to complement the ugui framework that came in 4.6.. ugui leaves a lot to be desired, kinda wish it was more like ongui now which still has the all encompassing guistyle and easier to build ui in code, ugui is a manual mess to deal with. unet i won't touch until I see what was meant to be phase2 and that fell off the radar. wont touch any 3d terrain stuff, just use meshes for environment now, could go on.. Unity is good for beginners, its great if you don't mind spending lots on the asset store or adjust your game ideas or how you do things. Though Unity does feel like a juggling act as far as performance goes. It's on going battle with whatever you're doing, its not perfect but the other engines aren't really any better, all things considered Though consider them you should depending on skillset, game plans, what you want to do learn etc

"Speaking of which, actual scene editor (window with transform gizmos and stuff) in unity is just plain horrible. Unreal engine has better transform widgets/controls within their scene editor and performs much better on large number of objects."

Unity's Maya like gizmo's are one of the nice things about it, apart from kinda crappy snapping, and no built in grid snapping options. Object selection on large number can be slow.. and just viewport camera flying leaves a lot to be desired both in glitchy pauses when switching directions .. and lack of options.. like to control speeds, even the shift multiplier.. i get so effing sick of it, either too fast, or too slow and if you stop it you have to wait for speed to pick up again.. then the weird quirks where you can zoom in/out with mousewheel.. and this also changes the camera clipping plane distances.. all completely under the hood with F*** all way too set things the way I want without messing with it. Terrible if you [F] frame on selection and depending on scale of that object, clipping and [wasd] speeds can feel completely off. These minor things where source code is great to just go change it upto the standard you want it at.

"Working with UI in a 3d scene is a pain. 2D ui exists in the same scene as 3d one AND 2d canvas blocks 3d objects from being picked."
heh opposite for me, 2d canvas blocks can't be quickly selected without background 3d objects being picked at times.

i don't mind 2d canvases in 3d scene.. but it sure is dumb that no canvas gui editor exist for the purpose of building ugui stuff, with proper adjustable snapping.. my workaround solution has been 2 scene tabs.. one 3d and the other is set in 2d with the canvas framed.. still messy, as far as scene management selecting background 3d object can still happen.

"Dragging object into scene always spawns it at some very random coordinates. (that seriously annoys me)."
After dragging into scene, you hold down [v] and move mouse to pivot location on the object (that is the stupid bit, that makes the entire process less snappy, because holding down [v] should start snapping the object immediately to other scene objects based on its default pivot.. in maya changing the pivot is another hotkey, unity combined it into one.. without thinking the entire process through.. they could keep it that way but then they should make it snap immediately if the mouse never goes over the object to adjust pivot location beforehand).. you drag pivot to snap onto other scene objects after setting the pivot down with [v] held down still.. it works but its kinda finicky.. and not as good as maya. Unity really should have carried on picking up more of maya controls and implementation on things, layers (show/hide, lock/unlock gameobjects based on set layer) etc

"- They are actually pretty active in the CryEngine Slack, but the forums themselves are pretty barren from both users and devs."

That's because there forums look fking awful, padding size per topic.. jeez just bad.. i dunno maybe they view there website in 4k at fullscreen and just forgot about designing S*** for the regular users.. bit like there engine .. they really should get the developers more active on the forum I don't care about slack.. but they have a fun a community show on twitch. I just feel CryEngine is little behind in features (catering towards indies anyway), but ok if you got the time to learn it or do a side project with it. The other area is doing any gui for a game.. is way too much effort than what it needs to be for indie's. Ie scaleform flash to do huds etc.

 

Yes. That were the only two nice features of C# I found to be useful after using C#.

It only means you are not familiar with the language.

See, every feature is there for a reason. If you're annoyed with headers (I miss having headers in C#, by the way), you don't understand the reason for the feature. Every single thing that is present in C++ is at some point useful. Even goto statement.

Also, I do not want to discuss that further.

 

Not wanting to critisize Epic too harshly (they have been quite busy), while Unity 5 has been a rough update cycle so far (things started to pick in the last few months, so maybe Unity is trying to make up for it), but:

If you look closely, there is plenty of feature/improvement requests that go unfullfilled until now. Forward renderer comes to mind. "Who still uses forward renderers?" Well, I guess that was Epics reaction to it. And just shows that they cater more to a very specific part of the devs, and is happy with patchy support for the others, whereas Unity tries to please everyone, and breaks down under the burden.

There are plenty of systems that are lacking as soon as you prod them too hard with nonstandard usecases (TXAA for example... that never worked for me. And Epics comment to other people not happy with the TXAA implementation was "we are still working on improving it". Good to know Epic is constantly trying to improve their engine. They probably shouldn't have kicked up so much dirt about TXAA if it wasn't ready for the prime time (same could be said about Enlighten in Unity).


The base tools in Unreal are quite good. A lot of things missing in Unity Stock editor, really. But then, that has been the "Unity philosophy" from the start, put a bare base engine there, make sure that base is solid, let the community fill in the rest. Which basically means if you want some of the more advanced tools, you need to take a trip to the asset store. And most of the time, pay additional money for it.
Yes, you pay additionally for something that is free in Unreal. But: from my expierience, while Unreals stock tools easely beat Unitys stock tools, they cannot compare to the functionality of some asset store tools. And you get premium support for that one asset on top of it (better than the "we are working on it" support from both Unity and Epic anyway).

For example, the Unreal spline tools are cool for many things. If you specifically are looking for tools to build a road system, I would dare to say some Unity Asset Store tools (EasyRoads3D for example) blow the bare bones spline tools in Unreal out of the water any day of the week. Given that the dev of this asset has sunken years into fintuning his asset for use with road creation that is hardly suprising.
But this and the additional support relativates the additional cost a little bit.
Sure, you can build your own system on top of the Unreal spline tools that match EasyRoads3D for features. But now your are just substituting paying some bucks in the Unity asset store for investing your own time to improve Unreals tools.


For me, Unreal cannot beat Unity on features if you look at the full package (and that for me includes community and Asset Store stuff)... given how long Unity has been in the business, and how active the community is, hardly surprising.
Difference is you get more in the basic free package with Unreal, to pay potentially more later. You have a higher upfront cost with Unity where you either need to invest your time to write or find the tools needed, or pay some money on the asset store for them.

It really comes down to liking the Unity or the Unreal way of doing things better, GameObjects vs Blueprints, C# vs C++, Upfront vs. deferred cost, easier for the beginner vs. more powerful for the freeloader.
I am still convinced Unreal will win some FPS on the performance front, but without a real fair comparison, I cannot make a statement there.
Features in Unity have never been a problem as long as you a) didn't mind tacked on systems, b) paying extra for extra functionality, c) were able to wait just a little bit longer for somebody to port it to Unity.


Personally I hope Unity will more and more concentrate on improving the core of their engine (like what they do with the instancing support), and let the asset devs handle creating convinient tools for game devs out of these basic functions.
It makes sure the engine is as fast and feature complete as possible on a low level, it makes sure we have some innovative solutions being developed fast and with good support, and after a year or two, Unity could still decide to port an asset store asset to their core engine so it also benefits devs that have no budget for it (hopefully paying the asset store dev a sum to make up for their lost business).

 

One comparison could be what job prospects the respective engines provide.

What companies and projects could you work on if you skill up in a specific engine?

 

Depends on the location. But here in Melbourne it's almost exclusively Unity. There area few larger/older studios running in house engines. So far I've only met one unreal developer, and she is working on her own.

This may be reflective of the Melbourne landscape. It's almost entirely built of small studios. Some of these studios punch well above their weight.

 

Unity is flat out better for small (tiny) teams and doesn't scale well for bigger teams (although you can fix this by rolling your own scene management and keeping a really tight lid on versioning).

I've got 1000 staff working somewhat indirectly on my game though, so there's that.

 

Would be nicer if they did exactly what I told them too.

 

I broke my left arm yesterday and I'm kinda appreciating Unity's mouse centric-ness today, since I can still be kinda productive in the editor with just one hand.

 

Good recovery.

 

See if you can train your OS to edit in Unity using voice control.

 

Speaking of which...

 

Very cool.

 

Brain Mice next step.... oh wait, we had that already and it sucked big time....


Mouse Centric is fine for dragging and dropping... not so much for navigating in a scene. Its usable, sure. But it drags your productivity to a crawl quickly. You only notice once you had a different kind of navigation for some time.
Its why I canceled my Maya LT sub quickly. Never groked how to navigate the viewport without a lot of mice action involved. Was a fine 3D modelling app else, but my productivity in Blender was doubled just because of the hotkey navigation and Space Navigator support.

Thus I really think neginfinity is right, we need a Blender like system of hotkeys in Unity to supplement the mouse centric viewport navigation. Don't remove anything, just give the user additional possibilities to define hotkeys (and maybe make that Space Navigator Plugin a built in thing, so I don't have to reopen the plugin window everytime my editor view crashes).
Would give the beginners the bare bone mouse navigation, and more advanced users could define their hotkeys, maybe synchronize them with their 3D Modelling app of choice so they don't have to learn 2 different sets of navigation hotkeys.

 

Since you like and know Blender shortcuts so well, why won't you make an editor extension adding support for them?

 

This. I thin blender like shortcuts are a horrible idea for Unity. But I think an asset that ads them to the editor is a clever business move.

 

How much are you paying for that, again?

Seriously, though... it may be a royal pain in the ***, because you'll need to rewrite transform gizmo and make sure that your version plays nice with default one.

 

Well, gotta say I have never even thought about it... you might be right, might be a big pain, but then again might be worth spending an hour just to see how much of a pain it really is.

Given that, if you keep this as a general "hotkey manager" that just listens to keypresses in the editor and exposes some internal Editor functions, it might not just be interesting for people that like the Blender shortcuts.
Everyone could define his own Setup. As long as it is hotkey based...

 

So how does UE4 compare, hotkey wise? It is incredibly trivial to code it in Unity while unity is still running. What about the competition?

 

To my knowledge Editor does not have api for blender-style axis movement locks, so you'll have to redefine all transform gizmos. There's "Tool" class for scene object manipulation, it is probably the class that will have to be used.

It is very likely to be doable, but it'll be like spending 2 or 3 weeks in order to get 4 sales on asset store.

It doesn't have blender hotkeys, but transformation widgets are better, grid is better, and it allows rapid level construction, and scene window is more pleasant to work with. One of the nice tricks is "shift drag" where you click object, press shift, drag and that creates object duplicate. You can very quickly fill large levels this way. You can also define new editor modes as described here:
http://blog.uppercut-games.com/post/92803859276/climb-system-introduction
Performance is superior. On a level I worked with, rectangle select of few hundred of objects in unity can freeze editor for half a minute (because repainting object wireframes is very slow in unity), while UE4 does not have this problem.

The difference is that Unreal 4 has different attribute towards prefabs. You see in unity "prefab" is an "instance" of class with custom data. In unreal 4 equivalent of prefab is a new "class" (with a name and all) which can be defined via blueprint or C++ class. Also, blueprinted objects do not participate in lightmap baking (that's a bug, though, which may be fixed one day). So, in unreal level, a static environmental mesh is treated as something akin to a pixel or primitive or polygon, while anything that has some purpose must be defined as a "class" (blueprinted or C++ one). That's just different workflow, though.

 

ctrl+d in unity?

I don't have this problem but I only picked around 400 objects. I do have a wild computer though so it seems like this could be optimised.

Regarding prefabs in Unity - this is what happens when you duplicate anyway, so I found that easier to wrap my head around than the UE4 approach.

 

Nope. Not the same thing. It is much faster.

The scene where I had this issue had 6 thousand objects in it. Horribly slow rectangle select in unity, no problem whatsoever in unreal 4. Basically, repainting "selected" wireframe was an obvious bottleneck.

 

1. D.
2. Click.
3. CTRL.
4. Drag.
5. Undo.
   
6. Click again because you clicked the original and not the dupe.
7. Wrong one again.
   
8. Fine, click it in the hierarchy instead.
   
9. Drag.
10. Repeat.

vs.

1. SHIFT.
2. Drag.
   
3. Repeat.

 

It automatically selects the clone though...

And you can intercept and code this in within 5 mins

 

I know, I just got up and I'm being dumb. So...

1. CTRL.
2. D.
3. Drag.
4. Repeat.

vs.

1. SHIFT.
2. Drag.
3. Repeat.

I still like SHIFT drag better.

 

Nu huh!!!


1) CTRL + D & Drag
2) Repeat

vs.


1) Place finger on Shift
2) Press Shift
3) Click & hold left mouse button
5) Move mouse slowly
6) Release finger
7) Exhale
8) Repeat

 

I don't pay anything for that, because I think it's a dumb idea (aside of per-axis locks so you won't accidentally move/rotate in the wrong direction).

He complains, he wants something that would do this, so why not develop it himself? I mean, Editor scripting is powerful and allows for virtually anything (even custom themes as it turns out) so why not make it happen? Complaining in the forums most likely won't make this happen, a couple of .cs files might.

 

FTFY

 

Same, I do it in max

 

1) Press alt by mistake
2) Look down in annoyance
3) press Ctrl
4) press F
5) Look down again
6) Press Ctrl
7) Press D
8) Fumble
9) HOLD Ctrl
10) Tap D
11) Realise it wasn't selected anyway

It's gone to level 11

 

That's Unity's attitude as well though. I mean it has a forward renderer but it is being neglected and a lot of things with forward/gamma are broken.

 

Well forward+ renderers are definitely coming back in fashion because they can be a lot faster, and still remain essential for mobile and VR...

 

I just want Unity's development to go faster. I'm sure Unity wants too, but for some reason it's not happening. I don't know what the issue is, but it's time they sort it out.

 

Well @tatoforever and I have been harping on about it over the years like a stuck record: http://forum.unity3d.com/threads/implements-forward-its-faster-than-deferred-rendering.132312/

 

Funny. Actually I checked it and in UE4 it is alt-drag and not shift drag (cause I was mostly working with unity last few weeks), but it goes like this.

1. Hold alt.
2. Start dragging the object. The moment you start dragging creates duplicate. While you drag, you drag duplicate.
3. Release duplicate in new position.

It is more convenient than ctrl+D-ing all the time.
------

Do you REALLY need such obvious thing explained to you?

It is because you don't pay anything for that.
Why should anyone spend their time increasing Unity's profits for free?

Developing a feature costs time. I can rewrite many of technologies from scratch myself, but the thing is I won't get squat from doing it, so in the time I'll be writing a feature I won't be getting any income to pay any of my bills, and when I'm done, asset store income (assuming I sell it) won't cover the time I wasted writing a feature.

Here's the thing, though. The company itself is in much better position for improving their product (they have source code access, they can skip asset store approval), and engine improvements increase their revenue. I'd expect them to have larger number of engineers that know engine ins an out better than anyone else and those people would implement missing features faster as well.

So, WHY should someone who wants a new feature spend their time implementing a feature when it is ultimately in company's interest to improve their product?

----

Frankly, "you want it - you implement it" attitude stinks and is one of the reason why opensource is not what it could've been. This approach encourages people to waste their time doing things inefficiently.

 

Yeah although who says such additions are right or even welcome? you like it. Others don't. Unity has to decide. If it's 50-50 it's not worth it, and it's better to instead create better documentation and make it easier to modify.

I mean, I come from a max background. Shift+Drag is the the way you copy or clone or instance there. I have never had a problem using ctrl+D in unity either. So I don't even believe this to be an actual issue over just making it easier for the individual to make personal changes.

 

Don't forget me here.! I've been telling Epic to sort one out since forever. I'll quote this from good old Ola Olsson himself who was kind enough to post on Aras's blog:

"
One thing that seems to go almost everyone by in this age of compute shaders and atomic operations is that this technique has been viable for years (for forward rendering)! It's only happenstance that it had to be discovered via the somewhat roundabout way of tiled deferred and compute shaders.

In my demo implementation I don't make use of any compute shaders or fancy GPU-built linked lists. It's implemented using OpenGL 3.3, but there are no reasons it couldn't be done using 2.0. Assuredly, it can be faster to do the tiling on the GPU, and, if using the depth buffer, it can avoid stalls, but it is still very quick to bin several hundred lights into a low-res grid using the CPU.

Just thought I'd point this out since everyone seems to connect it to modern GPU features of some description, and it may scare some people off from realising this may well be the absolutely easiest way to add many-light support to an existing forward rendering engine.

Cheers
.ola "

Going by what Ola says, seems it would be worth giving the FR some love.

@hippocoder

If nothing else Epic have been very smart about what features they implement, most 3D games will use a large portion of their toolset. Some will be slightly less relevant although extremely useful, but there's no real incentive for Unity in some cases.

Yeah quite a few asset store "features" are rubbish, but some of the bigger frameworks even if written by two or three guys are not only far in front of Unity they are in front of Unreal too.

Technically speaking Unity should have the man power and skillsets to literally trample Epic and any asset store vendor but going by past history in feature releases it looks kind of the other way around, but now they aren't behind paywalls and they have a bit of freedom let's see what happens.

In other cases, you can only have so many fingers in a pie and some of these assets on the store have been developed over years. Which you gotta kind of weigh up the pro's / cons on that one.

 

1000 staff now too! I think basically Unity is just warming up, and they do hire a lot of people who make asset store assets rather than buying the actual asset as this makes more sense on a number of levels.

 

This. Any focus on keyboard shorcuts in the blender style would be terrible for my workflow. Here is how doing anything in blender runs.

1. Select an object by clicking it
2. Curse at blenders backward clicking
3. Select the object by right clicking it
4. Press ctrl + shift + b + 16
5. Curse because blender didn't unselect a previous selection
6. Spam ctrl + z until I get back to the start
7. Google the short cut to deselect an object
8. Deselect the last object I used
9. Select the new object
10. Press ctrl + shift + b + 16
11. Curse because it's the wrong shortcut
12. Google the right short cut
13. Curse because my mouse doesn't have a middle button
14. Google how to simulate a middle button in blender
15. Press the right shortcut
16. Repeat

While I appreciate some users might enjoy this keyboard centric approach, it drives me mad. I prefer clean mouse based interfaces.

 

There are already assets for more keyboard centric use if you want that: Scene Navigator and SnazzyGrid2.

I don't begrudge needing/wanting to buy certain assets to make up for deficiencies in core Unity. Features I want(easy terrain creation) may not even be applicable for many devs. Also if I have multiple choices on the asset store I can pick and choose the one that works best for my project and workflow. The costs are pretty reasonable too. A $20 here, $60 there one time cost for an asset you may use for years is a solid deal. And the support is probably going to be better and a more personalized.

My only complaint is the lack of a "this isn't what I wanted/doesn't work for me" refund policy. I've ended up with some assets that don't work for me at all and have had to move onto other assets that do. Frankly the lack of a 24 hour refund window has probably kept me from buying a lot of assets.

I don't mind paying $60-100 for water or sky system, but I hate spending money on an asset only to see it's totally worthless for my use case and then being stuck with it.

 

For someone who's not familiar with blender, you mean.

Move object by 0.1 units.
1. Right click object.
2. GX0.1
3. Enter

Same in local space:
1. Right click the object.
2. GXX0.1
3. Enter

Basically, there are few controls you must know, once you grok them, you're set.

Spoiler

Space -> brings menu where you can type in commands
G - grab
S - scale
R - rotate
B - box select
X - delete
C - paint select
H - hide
Alt-H - unhide all
Shift+D - duplicate
Alt+D - make linked duplicate (clone of objects that shares all the changes with original).
Tab - Switch to/from edit mode.
Ctrl+Tab - switch object mode.
Z - toggle wireframe
Alt+Z - render using opengl/glsl (with realtime shadows from spots).
Shift+Z - render current viewport with rendering engines (cycles)

IN edit mode:
Ctrl+Tab - brings up mode selection menu.
Ctrl+Tab, 1 - edit vertices
Ctrl+Tab, 2 - edit edges
Ctrl+Tab, 3 - edit faces

Ctrl+R (In edge/face mode) - loop cut, mouse wheel to increase number of cuts.
K - knife tool
E - extrude
Ctrl+F bring up face menu
Ctrl+E bring up edge menu
Ctrl+V bring up vertex menu
w - bring up special menu
With verts selected: ctrl+F - make edge/face, depending on mode.
Two verts on face selected, J - join verts with edge
Alt+Rightclick - select connected elements (edge loops, face loops)

----

That's just a few

Basically, it is vim for 3d modeling. Hard to get hang of it, worth it when you figure it out. It is not keyboard-centric, because keyboard is used to rapidly change what mouse is doing.

 

Same. Personally I loved 3ds max (my high school had version 3.1) because it was largely mouse-based. Maya might be pretty good too but I haven't taken the time to actually try it out. I'm not an artist but I would love to give artwork a shot.

I'm seriously thinking of picking up a Wacom Intuos come next month when I have some funds to splurge.

 

First step is realising you really are an artist, it's a matter of trusting yourself, people go into it thinking far too negative. That's actually ordering yourself not to do it.

 

What I meant by "not an artist" was more along the lines of "I've never given it a chance" than "I can't do it". I know I can do it because I used to draw back in elementary school. I just largely stopped once I hit middle school and discovered how much I enjoyed programming. In retrospect I should have kept up with artwork too but that's hindsight for you.

 

We are using a custom forward renderer that does up to 3 dynamic lights with(out) shadows in one pass. It will be possible to convert it to a Forward+ and have more lights but as of now we didn't had the urgent need to do it as our game requires a small amount of light entities.

 

Yes, I totally acknowledge that if you use blender all the time then learning the hot keys isn't that bad.

I jump frequently between software packages. I work in engineering drawing software. I work with office a lot. I do a lot of programming in Visual Studio. I make games in Unity. And I occasionally use photoshop or blender for art.

Relearning interfaces each time I return to a program is a pain. And so far blender is consistently the hardest to get into after a break.

 

You can spend about $35 and have a laminated poster made of your favorite images of the Blender Keyboard with the Shortcuts written on the keys.

 

Hi guys!

Though I didn't want to ressurect this thread, I do believe that, for newcomers in both engines, the best answers are those explaining personal experiences, general benefits x drawbacks for each engine etc, which we have quite a few in this thread, but maybe just a few and a lot of derrails. Just wanted to add my experience and maybe a more updated version of things, giving that the engines are growing quite fast and the last answer was from the middle of the year.

I'm a game designer with a couple years of experience using Unity and less than a year using Unreal (which is my current choice so far). I'm by no means expert in any of those engines and I also just have a slight coding knowledge, I come from a 3D background and more recently focusing on mechanics and game design.

I'd like to start just pointing out that both engines are AWESOME and many people take that for granted - too much for granted. I used both and I love both - not to say that I didn't have my headaches with both.

Here's what I think each engine is NOT:

- Unity is for small games. I believe these days are gone. Each new update we see major steps in Unity. I'm currently playing Wasteland 2, a game made by some legendary devs at inXile Entertainment. Wasteland is a very complex CRPG and with maybe 100 hours of play. It is NOT a small game, like AT ALL, and the game's not even that recent, it was released in 2014, which means that it was being developed in 2012 at least, and it was made with Unity. You'd also want to check the more recent Project Wight. inXile is already making Wasteland 3 also in Unity and it looks gorgeous.

- Unreal is for triple-A games, it's bad for indies. That's also not true today. Though Unreal lacks the ease-of-use that Unity has, it's not that hard, it gets more and more intuitive when you learn and there are many small games being made with Unreal. See Solitairica, Dicetiny and The Siege and the Sandfox - yes, pixel art in Unreal!

- Unreal is too complex. That comes close, but it's not true in my opinion. Unreal has a more bloated structure, that is correct, but the thing is that everything is separated in editors and each editor is a lot to handle. But I've come from a 3D background and I know that complexity is there for you to use only when you need. Unreal is no more complex than Maya. In ZBrush, for example, a newcomer won't even know how to start modeling, LITERALLY. With time, you'll learn how to focus your attention on what matters in the editor. I'd be glad if they revamped things a bit though, I'm not gonna lie.

- C# vs C++. I have just a smaaaall knowledge on those, but I've read almost everything there is to read in the forums, especially Unreal forums where there's a good deal of people wanting C# integration (in fact, as far as I know there's already one unnofficial C# solution). What I can say is that there's absolutely no consensus, both languages are great in their own manner. Bottom line is that C++ is more perfomant but more complex. Tim Sweeney himself, Epic's founder, said that in this thread.

- Unreal is bad for 2d. That is in great part true. But it can be better put this way: Unreal is not intended for true 2d pipeline at the moment. Paper2D is a very good plugin, but so far without that much support. That may change in the future - or not. See this talk with the designer of The siege and the Sandfox and his team experience (triple-A veterans) using Unreal for a 2d pixel art game. Overall, I'd say if your intention is to focus on 2d games, especially if you know you'll be making only 2d games, there's not much sense in picking Unreal over Unity, but...

My Verdict:
1. Unity
Pros
- Easy to use and learn.
- Clean UI
- Fast to open and navigate
- Very good graphic fidelity
- Doesn't need to compile anything
- Very good starting features
- Asset Store is part of the Unity experience, and it's very good.
- C# is considered a cleaner and easier language for coding especially for indie games, which may not require that level of memory control and performance.

Cons
- Relies too much on the Asset Store and lacks a lot of features (no Material Editor, no Visual Scripting, no Cinematic tool, no good UI tools and a LOT more). This is controversial though; Unity intended from the start to give developers a great Store, and they did. Using Unity is knowing how to properly make use of the Store. It may be a great thing, but you'll be spending more to get some basic features.
- No source access. This is very important to help extend the engine especially in bigger and more demanding projects.


2. Unreal
Pros
- Complete package; you get Material Editor, robust animation tools, cinematic tools, particles... you name it. It's all in there.
- Top-notch graphics, particles, lighting, effects... and mostly out-of-the-box.
- Marketplace is VERY GOOD as well, but it's still growing. See these amazing Turn-Based Toolkit and CCG Toolkit. It's small compared to Unity's, but it's high quality and rapidly expanding.
- Blueprints. For me personally, the BIG WIN. As a designer, it's a great tool for me, and it's built to work with C++ more efficiently. The majooooority of people I see relating to BP talk about love or hate. But most people talk as if they were to work alone forever to make games. In a team with designers, artists, coders, engineers, the integration of BP and C++ for me is perfect. Unity has Playmaker and a couple other good solutions, but here you have to decide if you prefer 3rd parties or native tools.

Cons
- Sometimes working in Unreal is like playing Clash of Clans: YOU WAIT. With compiling all around, sometimes you just have to wait (a couple seconds or minutes, even to open the engine)
- I've found some annoying bugs more frequently than I'd like, but I argue that Unity has many others as well.
- Extra layer of complexity to beginners. That said, don't let that turn you away from UE4. It's great and it becomes more intuitive with time.

That's it, may not have covered it all, but did a huge text already.

Hope that was useful! Choose the engine which is more suitable for your mindset and your team, and don't be afraid. Unity rocks. Unreal rocks.